Spinal Cord Injury (SCI) leads to alterations in brain structure and function by spinal nerve damage, secondary inflammatory responses, and by the consequences of living with paralysis and neuropathic pain. Physical inactivity due to lower body paralysis rapidly leads to loss of muscle, and risk of heart disease. The leading cause of death after a spinal cord injury is cardiovascular disease, and just a year after injury, those with SCI have a peak exercise capacity half that of the unfit general population. The good news is that aerobic exercise reduces the risk of chronic metabolic and cardiorespiratory diseases, reduces inflammation and pain, and increases mood and quality of life. Exercise can also reduce brain inflammation, enhance endogenous analgesia, and increases the size of the hippocampus. The issue is that muscle paralysis in SCI restricts the ability to achieve the levels of exercise that is necessary for broad analgesic, anti-inflammatory and neuroprotective benefits. Arm exercise can have some effects on heart and lung capacity, but the small muscle mass is insufficient to produce more than modest aerobic work. With functional electrical stimulation (FES), leg muscles that are paralyzed can be made to contract, thereby allowing more of the body to be exercised. The full rowing stroke is produced by both the (stimulated) legs and arms, increasing the active muscle mass and resulting in an aerobic work-out that is intensive enough to improve heart, lung, and ? maybe ? brain function. In this clinical trial of sub-acute spinal cord injured subjects, we will study how 12 weeks of FES- RT, in comparisons to 12 weeks of wait-list, changes pain, brain structure, endogenous opioid function and brain inflammation. We will measure changes using positron emission tomography and magnetic resonance imaging We hypothesize a decrease in pain interference, an increase in hippocampal volume, increased endogenous opioid transmission in the periaqueductal gray, and decreased hippocampus neuroinflammation.